Trust Assessment
build-hook received a trust score of 14/100, placing it in the Untrusted category. This skill has significant security findings that require attention before use in production.
SkillShield's automated analysis identified 5 findings: 2 critical, 1 high, 2 medium, and 0 low severity. Key findings include Arbitrary command execution, Remote code execution: curl/wget pipe to shell, Broad Bash and Filesystem Permissions.
The analysis covered 4 layers: Manifest Analysis, Static Code Analysis, Dependency Graph, LLM Behavioral Safety. All layers scored 70 or above, reflecting consistent security practices.
Last analyzed on February 14, 2026 (commit 13146e6a). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings5
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| CRITICAL | Arbitrary command execution Remote code download piped to interpreter Review all shell execution calls. Ensure commands are static (not built from user input), use absolute paths, and are strictly necessary. Prefer library APIs over shell commands. | Manifest | skills/wpank/build-hook/SKILL.md:94 | |
| CRITICAL | Remote code execution: curl/wget pipe to shell Detected a pattern that downloads and immediately executes remote code. This is a primary malware delivery vector. Never pipe curl/wget output directly to a shell interpreter. | Static | skills/wpank/build-hook/SKILL.md:94 | |
| HIGH | Broad Bash and Filesystem Permissions The skill declares broad `Bash` permissions (`Bash(forge:*)`, `Bash(npm:*)`, `Bash(git:*)`) allowing execution of arbitrary commands prefixed with `forge`, `npm`, or `git`. Combined with extensive filesystem access (`Read`, `Write`, `Edit`, `Glob`, `Grep`), this creates a significant risk. A compromised `hook-builder` subagent could leverage these permissions to read, modify, or exfiltrate sensitive files, or execute arbitrary system commands beyond the intended scope of building a Uniswap V4 hook. For example, `forge script` can execute arbitrary Solidity code which can interact with the OS, and `npm` or `git` can be used for data exfiltration or further command execution. Restrict `Bash` permissions to specific, necessary commands (e.g., `Bash(forge:test)`, `Bash(forge:script)`, `Bash(forge:verify-contract)`) rather than wildcards. Re-evaluate the necessity of `npm:*` and `git:*` if they are not explicitly used. Limit filesystem access (`Read`, `Glob`, `Grep`) to only the directories and file types strictly required for the skill's operation. | LLM | SKILL.md:1 | |
| MEDIUM | Delegation to Unspecified Subagent with Broad Permissions The skill delegates its core functionality entirely to `Task(subagent_type:hook-builder)`. This subagent inherits all the broad permissions granted to the `build-hook` skill, including extensive filesystem access and wildcard `Bash` execution. The security posture of this skill is therefore entirely dependent on the trustworthiness and implementation of the `hook-builder` subagent, which is not specified or analyzed in this context. A compromised or malicious `hook-builder` could exploit the inherited permissions. Implement strict input validation and output sanitization for the `hook-builder` subagent. Consider sandboxing the `hook-builder` or granting it a more restricted set of permissions if possible. Ensure the `hook-builder` itself undergoes rigorous security review. | LLM | SKILL.md:47 | |
| MEDIUM | Insecure Foundry Installation Method The skill's error handling suggests installing Foundry using `curl -L https://foundry.paradigm.xyz | bash && foundryup`. This method executes a shell script directly downloaded from the internet without prior review, which is a significant supply chain risk. If the `foundry.paradigm.xyz` domain or the hosted script were compromised, it could lead to arbitrary code execution on the system running the agent. Recommend a more secure installation method, such as downloading a specific version of the Foundry installer, verifying its checksum, and then executing it, or using a package manager if available. Alternatively, instruct the user to manually inspect the script before execution. | LLM | SKILL.md:109 |
Scan History
Embed Code
[](https://skillshield.io/report/d90b21f6e1ac25a0)
Powered by SkillShield